RESEARCH PAPER

Challenges in re-designing operations and jobs to embody AIand robotics in services. Findings from a casein the hospitality industry

Erica Mingotto1& Federica Montaguti2 & Michele Tamma1

Received: 30 December 2019 /Accepted: 4 September 2020# The Author(s) 2020

AbstractThe adoption of artificial intelligence (AI) and service robots in the tourism industry for frontline service automation is generatinga growing interest. Although there is a fairly large body of literature about this research field, the impacts on the service encounterneed to be further investigated. The paper presents an action research project that led to employ the humanoid robot “Pepper”,equipped with a supervised machine-learning AI system, at the reception of an Italian hotel to provide information to clients. Thisallowed to explore the role played by this agent and the effects on the changing role taken by frontline employees (FLE) andcustomers. Findings show that this technology can act as an augmentation force and that FLEs’ role can evolvemainly into that ofenabler - of the customers and of technology -, innovator and coordinator, while customers may take above all the role of enablerof the technology. The study also contributes to introduce the new role of “AI supervisor” among FLEs.

Keywords AI . Service robots . Servicemanagement . Operations . Hospitality . Jobs

JEL classification O33

Introduction

The development of artificial intelligence (AI) and the en-hancement of robotics, together with a stronger digital con-nectivity, impact all business sectors, including services (Liet al. 2019; Makridakis 2017). Companies adopt these ad-vanced and smart technologies in order to improve operation

processes, minimize costs, enrich customer experiences orpropose new ones (Ivanov and Webster 2017). Indeed, thetechnological dimension can be considered one of the compo-nents of service innovation (Den Hertog et al. 2010). Thesetechnologies have become a common topic also in the traveland tourism industry, where they are generating a growinginterest and, at the same time, a kind of concern about thedisruptive effects they are having on customer experience,business processes and business models (Lam and Law 2019).

There is already a fairly large body of literature aboutthis research field, also regarding tourism. Studies haveinvestigated the phenomenon from several and interrelat-ed perspectives and seven main research domains can beidentified (Ivanov et al. (2019): 1) robot design (function-ality, mobility, appearance); 2) human domain, concernedwith both consumers’ and employees’ attitudes, reactions,roles, etc.; 3) robot manufacture (hardware and software);4) functions in tourism businesses, such as marketing,operations, etc.; 5) servicescape; 6) external environment,in terms of legal, economic, social etc. aspects; 7) educa-tion, training and research institutions.

Because of the rapid and continuous evolution of these tech-nologies and the complexity of the disruptive effects, the impact

This article is part of the Topical Collection on Artificial Intelligence (AI)and Robotics in Travel, Toursim and Leisure

Responsible Editor: Ulrike Gretzel

* Erica Mingottoericamin@unive.it

Federica Montagutifedem@unive.it

Michele Tammatamma@unive.it

1 Ca’ Foscari University – Department of Management, Cannaregio873, 30121, 0412348709 Venice, Italy

2 CISET - Ca’ Foscari University, Palazzo San Paolo - Riviera SantaMargherita 76, 31100 Treviso, Italy

Electronic Marketshttps://doi.org/10.1007/s12525-020-00439-y

on the way the service is provided and delivered - and alsoperceived by customers – remains one crucial issue to be furtherinvestigated. Further research contributions are needed in thefield of customers’ and employees’ attitude; of the changing rolesthat they may take in the service encounter; of service design; ofimpacts on processes, operations, jobs and organisationalredesigning; of employees’ training (De Keyser et al. 2019;Murphy et al. 2017). However, many of these research areascan be effectively investigated through field studies based onthe active involvement of companies and their staff, relying ontheir support and willingness to be studied and to share theirstrategies (Ivanov et al. 2019). Nevertheless, whilemany concep-tual papers have been produced in this area, empirical researchseem to be less developed and research methods such as actionresearch and projective techniques have not yet been applied(Ivanov et al. 2019).

The present paper would like to offer a contribution, par-ticularly answering the lack of empirical studies. Indeed, thestudy is based on a recent one-year action research projectcarried out by the authors in order to support the employmentof a humanoid social service robot, equipped with a super-vised machine-learning AI system, at the reception of anItalian resort with the task of providing information to clients.The project was not conceived as an experiment but was a realinvestment by the hotel and the software development businessinvolved. A constant field work side by side with the companymanagers and employees was implemented in order to supportthem in the adoption process. Regular observation of operations,meetings and interviews with the staff and with the clients wereset in order to monitor changes on operations and tasks. In par-ticular, the project allowed to explore the changing roles in ser-vice encounters that frontline employees (FLEs) and customersmay take as a result of the implementation of AI and servicerobots. This issue not only emerged to the authors as one onthe most felt questions by the partner companies and their stafffrom the beginning of the field work, but it has also been docu-mented in the literature by several conceptual studies as a crucialtopicwhen it comes to investigate the impacts on the processes offrontline services (Bowen 2016; De Keyser et al. 2019; IvanovandWebster 2017; Ivanov andWebster 2019c, 2019d; Larivièreet al. 2017). The paper introduces a role for FLEs not studied inthe present literature, i.e. the AI supervisor, a role of enabler oftechnologywhich not only is new, but appears to be a key one forFLEs access to other roles.

Literature review

Artificial intelligence and service robots in the traveland tourism industry

It is firstly important to provide an operating definition of AI,machine learning, service robot. McCarthy, who first

introduced the term “AI” in the 1950s, defines it as “the sci-ence and engineering of making intelligent machines, espe-cially intelligent computer programs” (McCarthy 2007, inTussyadiah 2020). AI refers to machines that simulate cogni-tive functions commonly associated with the human mind,since these machines are able to perform human-like tasks,emblematic of which is the ability to automatically and direct-ly learns from experience - machine learning (Aghaei et al.2012; Panch et al. 2018).Machine learning can be defined as asubset of AI, where algorithms learn associations of predictivepower from examples in given data (Panch et al. 2018). Thedistinction between supervised and unsupervised machinelearning was crucial for the present study. Supervised machinelearning – one of the most consolidated areas of AI and ap-plied in the case here presented, refers to computerprogrammes that learn relations between data inputs and out-puts, relying on the analysis of outputs of interest defined by ahuman supervisor. Unsupervised machine learning does notrequire external definition of associations of (Panch et al.2018). However, at the moment most AI technologies arebased on “narrow” and “targeted” machine learning systemsthat perform in a quite constrained environment (Panch et al.2018), as in the case discussed in this paper.

Robots are defined as physical devices that are able toperform specific tasks, thanks to a certain degree of autonomy,mobility and sensory capabilities (Murphy et al. 2017; Chenand Hu 2013). They are distinguished between industrial ro-bots and social service robots, the one adopted in this study.While industrial robots are employed for performing industrialtasks, social service robots are designed to support and servicehumans by interacting with them and the environment in anatural, user-friendly and socially-compatible way (Ivanovet al. 2017; Pinillos et al. 2016). As a consequence, theymay be used also as the interaction counterpart of customersduring service encounters (Wirtz et al. 2018; Kuo et al. 2017).Service robots can be differentiated according to several attri-butes: their representation (physical robot, as Pepper, or virtu-al, as Alexa), anthropomorphism (humanoid or non-humanoidrobot) and the tasks they can perform (cognitive-analyticaltasks and emotional-social tasks) (De Keyser et al. 2019;Wirtz et al. 2018). The efficacy of a social service robot de-pends on three levels of development: hardware, i.e. the bodyof the machine, its sensors and the movement system(actuators); functionality, i.e. the software which enables func-tions such as visual and voice recognition, dialogue, mobility,knowledge representation ofmindmodels; and service, i.e. thetasks that the business assigns to the robot in order to createand deliver added value to the customers (Kuo et al. 2017).

Considering in particular robots employed in service andtourism companies, their acceptance by the customers and thehuman-machine interaction positively depend on a combina-tion of factors. Some of these refer to the user himself, such ashis attitude towards machines, his acknowledgement of their

E. Mingotto et al.

advantages, his past experience with technologies (Ivanovet al. 2018). Other factors refer to the robot, i.e. its embodi-ment and its anthropomorphic characteristics, perceived intel-ligence and perceived security (Ivanov et al. 2018; Murphyet al. 2019; Qiu et al. 2019; Tung and Au 2018; Tung and Law2017; Tussyadiah and Park 2018). Consumers are more recep-tive towards robots with animated features but they could benot always inclined towards humanlike robots (Yu 2019).Indeed, the Uncanny Valley theory (Mori 1970) suggests thatas robots become more human-like, users’ reaction tend to bepositive until they perceive them as too bizarre, weird orcreepy. The customer-employee rapport building is crucialfor mediating the relationship between robot attributes andconsumers’ reaction (Qiu et al. 2019) and the kind of tasksperformed by the robot is an important factor in customers’acceptance (Ivanov et al. 2018). Greater approval is shown forhousekeeping activities, information provision, processing ofbookings, payments and documents, and item delivery(Ivanov and Webster 2019a, 2019b; Ivanov et al. 2018).

The rapid technological evolution is leading also in theservice industry to the development of applications showinggreat potential that would have been unthinkable until recently(Ivanov et al. 2017; Wirtz et al. 2018). Although the diffusionof these advanced technologies in the tourism industry is stilllow (Ivanov et al. 2019), machines powered by AI and roboticsystems have started to be implemented in recent years bytourism businesses (from hospitality to airports, from airlinecompanies to restaurants, from museums and amusementparks to information centres) in multiple daily operations inorder to support service innovation and automation (Collinset al. 2017; Ivanov 2019; Ivanov and Webster 2017; Ivanovet al. 2017; Li et al. 2019; Makridakis 2017; Tung and Law2017). Businesses have been automating operations for a longtime in order to increase productivity and cost efficiency(Buhalis and Main 1998; Ivanov et al. 2017; Lam and Law2019). However, the potential of AI, digitization and roboticsmay further improve and accelerate the automation of process-es traditionally owned by FLEs, by involving new areas ofbusiness operations and then significantly altering the inter-play between the organization and the customers (Larivièreet al. 2017).

The reasons why tourism businesses look with great inter-est to the adoption of AI and robotics are the opportunity: a) toimprove efficacy and efficiency, since these technologiescould work longer than humans and perform their work morecorrectly and in a timely manner (Dogan and Vatan 2019;Ivanov 2019; Ivanov and Webster 2017); b) to save labourcosts, substituting or supporting employees, relieving themfrom manual, repetitive and tedious tasks and allowing themto dedicate to more challenging and creative activities; c) toimprove service quality and customers’ experiences, by intro-ducing new attractive, funny and engaging ways for deliver-ing services (Ivanov 2019; Ivanov and Webster 2017).

Regarding this last aspect, some researchers suggest that insome cases consumers tend to co-create new experiences, byproactively interacting with robots for developing a certainlevel of “relationship” with them (Tung and Au 2018).

However, the application of systems based onAI and socialrobots has to be limited for the moment to quite structuredsettings and it implies some significant costs, which oftenare not considered enough by tourism businesses: on the onehand, financial costs linked to acquisition, installation, main-tenance and update of the technology and to the hiring ofspecialists to manage it and to the training of the staff; onthe other, non-financial costs, in particular those connectedto the resistance of employees (Ivanov 2019; Ivanov andWebster 2017), who could be frightened and psychologicallystressed (Li et al. 2019) by these disruptive changes and couldrefuse to use technology. Together with employees’ resis-tance, implicit costs can also arise from customers’ resistance(Ivanov 2019; Ivanov and Webster 2017). Although techno-logical innovation is adopted to improve customer experience,this outcome cannot be taken for granted since customersexpect some degree of effort to learn how to use technologicaldevices and, if this is perceived too difficult, they might giveup (Ciuchita et al. 2019).

New emerging technologies, such as AI and service robots,have great potential in terms of productivity improvement,service quality and development of new consumers’ experi-ence; however, they are not yet perfect and capable of doingeverything. Companies are required to adopt them carefully,by making an in-depth evaluation of benefits and costs. Anextreme resistance towards these advanced technologies butalso an excessive enthusiasm are both to be avoided (Ivanovand Webster 2019a, 2019b, 2019c, 2019d).

The technology adoption process in service andtourism companies

The adoption of systems based on AI and social robots maybring significant challenges to service and tourism businesses,as it modifies the ways in which the service is delivered andexperienced, and thus the relationship between customers andproviders and even the roles of employees (and in particularFLEs) and customers (Bolton et al. 2018; De Keyser et al.2019; Ivanov 2019; Ivanov and Webster 2019c; Larivièreet al. 2017; Li et al. 2019). van Doorn et al. (2017), discussingthe technology infusion in frontline experiences, introduce thenew concept of “automated social presence” (ASP), referringto the ability of the technology to effectively engage con-sumers in significant social encounters. In this sense, the tech-nology able to act on this social level becomes a real socialagent, like FLEs, other staff members and/or other customers.

Since these technologies can change the way companiesoperate, compete and cooperate (Ivanov 2019) and affect bothemployees and customers (Ivanov and Webster 2019c), a

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

crucial issue to investigate is how companies can effectivelymanage the adoption process. The literature review seems toshow that relatively few empirical studies focus on automationand innovation in the service and tourism industry throughnew advanced technologies (not only AI and robots), withparticular reference to the adoption processes undertaken bythe company and to the implications on the service encounter.

Some studies, which investigated the implementation ofradio frequency identification (Cobos et al. 2016) or ofCRM (Cruz-Jesus et al. 2019) in the service companies froman organizational perspective, explain that the technologyadoption process goes through several steps, such as knowl-edge, persuasion/evaluation, decision, implementation andconfirmation/routinization.

Other studies, which explored technology adoption formarketing automation, underline that, when a new technologyis included in the business environment, managing the chang-es on the organisational processes is crucial in order to ensurethe automation success (Järvinen and Taiminen 2016;Murphy2018; Speier and Venkatesh 2002).

Focusing in particular on service innovation and automa-tion through AI and service robots, the literature underlinesthat the implementation of these innovative technologies re-quires an in-depth analysis of costs and benefits and of theservice delivery processes; the re-engineering of entire activ-ities, processes and structures; the alignment of strategies andresources such as people (including customers), mind-set andleadership, so that the technological toolset is embedded intoday-to day operations and becomes part of the organization(Ivanov and Webster 2017; Ivanov and Webster 2019a,2019b, 2019c, 2019d; Kuo et al. 2017; Lam and Law 2019).

The adoption process of such kind of digital and techno-logical systems requires efforts both on the part of managersand on the part of employees, since a common understandingof the new organizational context is needed in order for theright skill-sets and mind-sets to be developed and new revisedoperations and practices to be put in place. It is a complexorganizational learning situation, in which people have tolearn new skills to face new challenges but, first of all, haveto be aware that they need more knowledge and be ready tolearn (Lam and Law 2019).

One of the most debated topics about the effects of roboticsand AI technologies on organizations is the replacement ofhuman workforce and, in connection to this, the changes injob profiles and in the well-established working relationshipsamong employees and between employees and managers (Liet al. 2019). The introduction of these machines may nega-tively affect the employees’ sense of workplace belonging anddedication, unless organizational support is provided, for in-stance through internal training programs, including soft skillsdevelopment (Li et al. 2019).

Since different combinations of automated and human so-cial presence in the frontlines can be configured (van Doorn

et al. 2017), managing the adoption process means firstly todefine the roles that technology can play in the service en-counter. It emerged that it does not necessarily have a “sub-stitution” effect. Indeed, within the conceptual frameworksdeveloped by Marinova et al. 2017, Larivière et al. 2017 andthen by De Keyser et al. 2019, technology can act as a substi-tution force, in particular when FLEs no longer take an activepart in the service encounter; it can be assumed that moderntechnologies may, in a way, also replace the customer, bymaking some decisions on his behalf (De Keyser et al. 2019;Larivière et al. 2017; Marinova et al. 2017). However, tech-nology can also act as an augmentation force when it assistsand complements human actors – both employees and cus-tomers - to better perform their tasks in the service encounter(De Keyser et al. 2019; Larivière et al. 2017; Marinova et al.2017), or it can become a network facilitatorwhen it is used toconnect different human and technological entities in the ser-vice encounter (Larivière et al. 2017). The augmentation andsubstitution force have been clearly studied and attributed tosocial robots and AI also by Ivanov and Wester (Ivanov andWebster 2019d).

Another important question concerns what new roles FLEsand customers will play. Indeed, as a result of the roleassigned to the technology, both employees and customersmay take one or more transformed roles. The frameworksdeveloped by Bowen (2016) and Larivière et al. (2017) ex-plain that they can become: enabler (for example whenemployees support customers in using the technology;or when customers, with their input, enable the technol-ogy to perform its tasks); innovator (when they takepart in the development and delivery of new services);coordinator (when they are a leading party in managingthe relations between the different networks and chan-nels); differentiator (when they directly contribute tocustomizing the technology and add it unique dimensionto). However, if employees and customers are not madeaware and thus do not perceive the benefits ofperforming their new role, they might not perform asneeded (Bowen 2016; Lar iv iè re e t a l . 2017) .Investigating in particular the adoption of physical andvirtual conversational agents in frontline services, DeKeyser et al. (2019) underline that more in-depth re-search is needed in order to investigate which new roleswill be dominant for customers and FLEs, also whenconversational agents do not substitute employees butwork together with them.

During the action research project carried out by the au-thors the main issues and questions that emerged were of asimilar kind, namely the changing role of FLEs and cus-tomers, also in relation to the role that the technologywould have taken. Therefore, the present paper developsits research questions starting from the last body ofliterature discussed above about the changing role of

E. Mingotto et al.

employees and customers when frontline technologies,and in particular conversational agents empowered byAI and robots, are adopted in tourism companies.

Methodology

The project context and themain technical features ofthe robot and of the AI training

The study develops within a one-year project carried out bythe authors from June 2017 to June 2018, aimed at the adop-tion by an Italian resort of a humanoid social service robot (therobot “Pepper”) equipped with a supervised machine-learningAI system. The resort –which includes two hotels for a total of400 rooms, one residence, and several services such as a well-ness centre, two restaurants, four swimming pools, a golfcourse, etc. - is part of a small Italian hotel chain and is locatedin one of the most popular Italian destinations for leisure tour-ism by the Garda Lake, in Northern Italy.

The robot was meant to be employed at the reception of theresort for answering guests’ simple questions, namely by pro-viding information about the hotel services and the destina-tion, in Italian, English and German (i.e. the main languagesspoken by the guests of the resort). The decision to adopt therobot was made by the CEO and the managers of the hotelchain in order to provide guests with another touch-point forasking for information in addition to the FLEs. The latter notonly had to check guests in and out, but were also continuous-ly pressed to answer crowds of guests with different needs andrequests related to the large number of services offered insideand outside the resort. The aim of the robot adoption was toallow guests to easily and quickly obtain information on sim-ple issues (opening hours, costs, location, of the services)without wasting time, and, at the same time, to “emancipate”FLEs from the repetitive task of answering such questions,and give them time for check-ins/outs, thick/rich interactionsand back office work.

The CEO and the managers of the hotel chain and thebusiness in charge of developing the software for the robotand training the AI asked the authors for a scientific support inintroducing the new technology and in managing the changesin service encounters and in operations. This, also becauseboth businesses were investing a lot in terms of equipmentand work for this innovation, which nobody had previousexperience of. Furthermore, this innovation was, for the soft-ware developer, the first step in launching a new business,and, for the hotel chain, the first opportunity in re-organisingsome processes and re-designing the servicescape. For bothcompanies the risk was high and the competitive advantage tobe obtained crucial, so they accepted to negotiate access toinformation, procedures, employees, to receive the help theyfelt they needed in managing the change.

This immediately appeared to the authors as a researchopportunity, and specifically for the investigation of the im-pact of such technology on frontline processes and jobs. Thisstudy is probably an unicum. Although at the time some socialrobots and AI were starting to be employed at the front officeand in other jobs in the service industry, none of these firstexperiments, to the best of the authors’ knowledge, involvedan academic approach, and specifically an academic approachdifferent from ones centred on computer science and IT de-velopment. Conversely, in case of experiments lead by mem-bers of academia the companies were used as an environmentfor testing the AI/robot, but had no a real commitment/engagement in the project.

The robot was chosen by the CEO and managers of thehotel company, with the advice of the software developer. Itwas manufactured by SoftBank Robotics and equipped withmicrophones, cameras, sensors, gyroscope, wheels, wi-fi con-nection and a touch screen. It is optimized for human interac-tion and it is able to a) speak; b) recognize sounds and voices,images, objects and faces; c) analyse the voice tone; d) moveits head, arms and back and move in all directions. The inter-action with customers occurs mainly through voice recogni-tion and conversation, although some information (such asmaps, weather conditions) is also displayed on its touchscreen.

The software developers programmed the robot, selectedan external AI platform and developed the interface betweenthe robot software and the AI. TheAIwas based on supervisedmachine learning and therefore required solid informationcontent as input in order to start learning and later a regularsupervision by a human, who has to check if the AI is under-standing and answering correctly and otherwise provide itwith correct inputs, and to update the information.Additionally, the AI learns and performs only for the specifictask it was trained for. The preliminary information contentwas developed by the authors by identifying, with the help ofthe FLEs, the most frequent questions asked by guests and thepossible answers. A set of 280 information items (composedof question and answer) was derived and translated in each ofthe three languages. The basis for the initial AI training wasthe same for the three languages. However, when the robotbegan to interact with customers, the information content ofeach language started to automatically grow with differentitems, depending on the questions asked by the guests.

Research questions

The adoption of new advanced technologies - in particular AIand social robots - in a service environment is a complexprocess. As documented in the literature by several conceptualstudies, when it comes to investigate the impact on frontlineservice processes, an important issue is related to the roletaken by the technology and to the changing role of FLEs

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

and customers. Indeed, it is strictly connected to other strate-gic and organizational issues, such as: the reengineering ofoperations; the redefinition of tasks and job profiles; the de-velopment of new skills both by employees and clients; theservice encounter redesign (Bowen 2016; De Keyser et al.2019; Ivanov and Webster 2017; Ivanov and Webster2019c, 2019d; Larivière et al. 2017).

This emerged as one of the most felt issues by managersand FLEs also in the project under consideration, immediatelyfrom the beginning of the field work. The technology, as anew agent in the organisation, could have substituted or en-hanced some human jobs or some specific tasks; it could havemodified in some way the interaction with consumers. In ad-dition, even if based on advanced and smart systems, the tech-nology would have needed to be managed and supervised byhumans. It then emerged that some operations, tasks and rolesboth of employees and customers could have been changed;the question was how they would have changed and whatimplications this would have had for the organisation, forexample in terms of redesigning operations, developing newskills for employees or “educating” customers. This issue thenled to derive the research purpose and the research questionsof this study.

Three research questions were identified, starting from theconceptual framework theorised by Bowen (2016); De Keyseret al. (2019), Larivière et al. (2017).

1 Which role do these technologies play in the organization:augmentation, substitution or network facilitator? Therewas the need to investigate and clarify, at the beginningof the study, the expectations of the hotel company to-wards the conversational agent and, after its introduction,to what extent this agent was able to act as expected oreven better.

2 With respect to frontline employees: a) which changingroles can they take among enabler, innovator, coordinatorand differentiator? b) towards whom: the technology and/or the customer?

3 Which changing roles can customers take among enabler,innovator, coordinator and differentiator?

Research design

Given the nature of the project, the study was based on aqualitative and descriptive approach and in particular on theaction research methodology. Action research is the system-ized collection and analysis of data aimed at generatingknowledge that enables actions for making change, and ismainly based on focus groups, participant observation andfield notes, interviews, diaries and personal logs, question-naires, and surveys (MacDonald 2012). Action research canbe considered a valid tool particularly for operation

management, since it allows to face management and opera-tional challenges while simultaneously contributing to knowl-edge improvement (Coughlan and Coghlan 2002). It was thenthought particularly suitable for the present study, since therewas the need to balance the research purpose with the requestsof both partner companies, who needed a consultancy support.

The tools implemented for this action research were: 1)semi-structured interviews and meetings with managers and/or executives; 2) analysis of the operations manuals in usebefore and after the implementation of the technology, andfield observations of the actual procedures followed duringthe FO encounters; 3) participant observation of customers;4) interviews with customers; 5) analysis of the data recordedby the AI platform during its interaction with customers.

Figure 1 shows the conceptual framework and Table 1a, bdescribes the main phases of the overall project, the mainactivities specifically carried out by the authors for the actionresearch and the tools implemented.

With respect to tool 1), meetings and semi-structured inter-views involved 3 employees at the company corporate level(the CEO of the hotel chain, the communication and the busi-ness development managers); 7 employees of the resort (theGM - General manager, the FO - front office - manager, and 5FLEs). Interviews and meetings were carried out face-to-faceand moderated by researchers. The participants’ statementsand answers were immediately transcribed and then collatedand analysed, as in Cobos et al. (2016) and in Lam and Law(2019). Firstly, all transcripts were read, in order to understandtheir overall meaning; secondly, significant statements regard-ing respondents’ opinions and beliefs were extracted; finally,the most common important themes were identified. Meetingsand interviews were carried out prior to and following the finaladoption of the AI. Before the implementation, topics inves-tigated were: a) the main characteristics of the front office, interms of main processes, operations and tasks (Cobos et al.2016); b) managers’ goals and expectations and frontline em-ployees’ attitude toward the robot and the AI. After the AI androbot adoption: c) if and how FO tasks and operations werechanged or were expected to change; d) employees’ trainingneeds in managing the robot, supervising the AI machinelearning and in supporting the customers-machine interaction(De Keyser et al. 2019). Twelve sessions of interviews andmeetings were carried out for a total of 40 h, 27 of which afterthe robot came into operation.

Tool 2) was the analysis and comparison of old and newoperations manuals used in the resort before and after the AIadoption, focussing on the description of the main procedures,tasks and job profiles of the FLEs. In conformity with the actionresearch approach adopted, the authors developed, togetherwith the director and the FOmanager, new operations manuals.

Regarding tool 3), participant observations were used toinvestigate customers’ interaction with the robot. The obser-vation protocol was designed on the basis of previous studies

E. Mingotto et al.

available in the literature about the application of participatoryobservation for the assessment of a service robot in a hotelenvironment (Pinillos et al. 2016) and for investigative pur-poses in other fields (Escuer et al. 2014; Yalowitz andBronnenkant 2009). The protocol was based on the pen-and-paper system, which prescribes to unobtrusively observeguests and to manually record data on their behaviour. Fourcategories of variables were chosen to be observed in order todescribe and contextualize how guests approach and interactwith the robot: a) stopping behaviour: total time of the inter-action (minutes and seconds); 2) interaction behaviour: mo-dality (mere observation of the robot, verbal communication,use of the tablet, touching of the robot, other gestures); inten-sity (number of questions or other kinds of sentences ad-dressed to the robot); perceived emotions (neutrality; surprise;joy; amazement; embarrassment; fear; anger; contempt) (cf.Tussyadiah and Park (2018); 3) situational variables (time,noise, crowding); 4) demographic variables (estimated age,gender, number of adults and children in the party). 22 one-hour sessions were carried out by the authors in different days

during the two months after the robot introduction and 65guests were observed, with an average interaction durationof 1:42 min.

With regard to tool 4), the interviews with customersaimed to integrate the findings from the observations. For thisreason, the decision was taken to include in the interviews theguests previously observed and to personally interview them– for about 10 min - shortly after such observation. The ques-tionnaire included mainly closed questions based on a Likertscale from 1 to 7 about the following information categoriesderived from previous studies about consumers’ attitudes to-wards robots (Ivanov and Webster 2019a, 2019b; Ivanovet al. 2018; Kazandzhieva and Filipova 2019; Murphy et al.2019; Qiu et al. 2019; Tung and Au 2018; Tung and Law2017; Tussyadiah and Park 2018): 1) customers’ attitude andwillingness to talk with a machine and situations in whichcustomers prefer to be assisted by the robot and by humans;2) user experience (ease of interaction with the robot, satis-faction about the answers received, robot appearance, etc.); 3)level of customers’ awareness about the robot/AI and their

Table 1 a. Phases of the overall project b. Activities and tools of the action research carried out during the project

a.

PhaseMonths

Jun17 Jul17 Aug17 Sep17 Oct17 Nov17 Dic17 Jen18 Feb18 Mar18 Apr18 May18 Jun18

Phase 0

Knowledge among the actors involved:

hotel company, software house,

researchers/authors

Phase 1

Understanding the context and

preparing the conditions for the robot

development and AI training

Phase 2

AI training and implementation of the

robot interaction functions

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

past experiences with such technology. Among the guestsobserved, 30 customers completed the questionnaire (12 inItalian and 18 in English), i.e. all those who had accepted tobe interviewed.

For a better understanding of the customer-machine interac-tion, attentionwas paid also to data automatically recorded by theAI during the human-robot interaction and displayed on the userinterface (tool 5). Indeed, the AI textually records all interactionsbetween robot and guest, intended as conversations in which theguest pronounces one or more questions or other completesentences and the robot replies. The AI user interface was setup for displaying all the conversations and their content and forproviding summary indicators, such as the number and averagelength of interactions and the topics of the conversations in as-sociationwith their occurrence frequency (for example restaurantopening hours, location of the swimming pool, or location of thecity centre). This last kind of information was considered inparticular in order to identify the most frequent questions askedby customers to the robot during the 41.600 interactions recordedby the AI in the two months after its introduction.

Findings

The role of technology

As for the first research question, i.e. the role of technology,the preliminary interviews and meetings with the CEO andmanagers highlighted that they expected that the robot couldprovide basic information to the guests, while the FLEs wouldhave more time to continue assisting guests for more complexservices and thick/rich interactions with customers andperforming check-in/out and back-office operation. Theyintended the technology above all as an enhancement of hu-man jobs, although they did not rule out that the machinecould eventually perform other frontline tasks in more or lessshort time, thus partially substituting FLEs.

Findings from the analysis carried out after the robot imple-mentation (from April to June 2018) seem to confirm that thetechnology has gradually started to act as an augmentationforce. In the first few weeks, 41.15% of the guests observedappeared confused and agitated: they were not able to clearly

Table 1 (continued)

E. Mingotto et al.

understand what the robot’s functions were and they mistook itfor a gadget; they found it difficult to start the interaction be-cause they had to select the language on the robot touch screenand, according to the data recorded by the AI, they asked non-coherent questions the AI had not been trained for (such as“how old are you?”; “can you dance or sing?”; “can I take apicture with you?”). However, in the following weeks, guestsgradually started to recognize the robot as an assistant. Thischange was the result mainly of two factors: firstly, the AIbecame more effective in providing correct answers, thanks toits gradual improvement through its machine-learning (fromabout 51.57% of exact answers provided by the AI in the firstweeks of April to more than 75.33% of exact answers at the endof June); secondly, guidance about the robot functions wereprovided to clients through information panels, and FLEsstarted to introduce the robot to guests during the check-inoperations and to invite them to use it for basic requests ofinformation. Findings showed: an increase in numbers of cus-tomers who interact with the robot in a correct and smooth way(from 58.85% to 88.24% of guests observed); an increase ofconsistent questions asked by guests - from 32.96% to 52.14%;an increase in customers’ satisfaction, recorded through ques-tionnaires, from an average value of 1.83 to 4.10 on a scale of 1(minum) -7 (maximum), and customers’ willingness to useagain the robot (from an average value of 1.70 to 4.06).

Findings showed that the conversational agent is able to fulfilits task, i.e. supporting and working in conjunction with FLEs soas to improve the quality of the service encounter. The technol-ogy here implemented, if further programmed, has the potentialto perform other tasks as well (for example check-in/out), but atpresent it is still too limited to engage in more complex interac-tions with the guests and entirely substitute FLEs.

The changing role of employees

As for the second research question, i.e. the evolving role ofthe FLEs, the comparison between operation processes beforeand after the robot introduction clearly highlights how theFLEs’ role has changed, and particularly the role and tasksof some among the FO executives chosen to be the AI trainingsupervisors.

Enabler towards customer and technology

Before the beginning of the project FLEs had no role insupporting customers in using technology. The analysis of theprocedures and the interviews explained that no technology,except the basic back-office PMS, was used specifically forservice encounters. This is not to say the employees and man-agers did not use any technology in their work. The hotel wasprovided with a typical infrastructure including a PMS, channelmanagers, revenue management systems, etc. However, thesetools were not meant for the customer to use directly or as amediation in the relationship between the customer and thefront office employees or any other relationship between thehotel personnel and the customers. The only technology,website aside, developed for the customers was an in-houseapp, with respect to which, however, receptionists were notspecifically required to provide assistance, andwhich the clientsdid not like to use. At the hotel, service frontline experienceswere therefore low in automated social presence and high inhuman social presence (van Doorn et al. 2017).

Contemporary to the introduction of the robot, the inter-views with the managers and the FLEs highlighted the needfor a specific training on the basics of the AI/robot use and its

Fig. 1 Conceptual framework(adaptation on De Keyser et al.2019, Larivière et al. (2017))

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

limits. A specific training was devised and delivered by theauthors, not only on technical characteristics but also on howto interact with it the robot, the kind of information the AIcould provide, and the need for it to be further trained andmonitored. FLEs were specifically encouraged to interact withthe robot themselves and to ask clients to do the same.Together with the GM and the FO manager, a specific entryon interaction procedure and its technical basics was added tothe FO operation manual. Interviews showed how this trainingwas felt as essential by the software developers, to ensure aneffective and autonomous functioning of the robot and AI.Not only the developers underlined the need for some training,but specifically for a training centered on involving FLEs roletowards the robot, the AI and the customers, more than ontechnical aspects (how to switch it on, clean it, check its wi-fi connection or audio system, how to access the API andcheck the training, etc.). The developers clearly stated thatthey realized that the key aspect to make the project a successwas not only for the technology to run smoothly, but also toprovide the maximum attention to the AI training within thehotel, and to set the conditions to encourage customers to useit as much as possible. It has to be recalled that, at the time thestudy was conducted, this was the first implementation for thesoftware developers, and one of the first in general, so a failurein the activity of this robot/AI) would have compromised awhole new possibility for business development for them. Thehotel management, since the project implied the introductionof an advanced technology in the area of service encounterswhere there was none before, agreed on it. The analyses of theprocesses and procedures before the AI introduction and theknowledge of how the robot/AI worked, led the authors to asimilar conclusion, and all the parties agreed on the need forspecific training.

This being more their expertise area, the authors were themain responsible for the design and delivery of the training,but content and approach were all agreed upon and developedtogether with the parties involved.

When the robot had beenworking for a month, it emerged aneed to further reinforce the “message” provided during thetraining to FLEs. The supervision of the hotel managers onhow the customers were reacting and interacting with therobot/AI led them to conclude that there could be improve-ments in the level of engagement and interaction between therobot and the customers.

The developers and researchers’ check on the AI workconfirmed that it was answering correctly to the majority ofthe questions (from about 51.57% of exact answers providedby the AI in the first weeks of April to more than 75.33% ofexact answers at the end of June) but the interactions wereshort and superficial. On this basis, further interviews withthe FLEs and managers, in combination with the participantobservation, led the researchers, the FO manager and the GMto agree that the new role of the FLEs had to be made more

apparent - for themselves and for their colleagues - and that anintervention on the procedures could effectively encourage thecustomers and compel the employees to become enablers to-wards the technology. Specifically, it was decided to changethe existing check-in procedure, and, more in detail, to shortenthe information about the hotel given during the final steps ofthe check-in, instead introducing a sentence that invited thetourist to “ask for further information you may need later andin the next days to Robby, our robot concierge”.

The training, the redesign of the procedures and then of themanuals clearly highlighted the role of the FLEs as enablers,who support the customers on how to interact with andwhat toask to the AI/robot. This confirms the initial hypothesis thatthe introduction of the AI would lead the FLEs to assume anenabler role, as also supposed by Larivière et al. (2017).

In addition, findings highlighted a far more importantchange in the role of some of the FLEs. A completely newone emerged: the “AI supervisor”. The need to create this newrole came from a specific request of the ICT company thatworked on the interface between the AI and the robot soft-ware, as the software developers underlined that, for the sus-tainability of the project, there was the need to identify some-one within the hotel staff who would be in charge of trainingthe AI in the future, since the developers did not plan to do thisthemselves and also stressed how this would speed up theprocess in any case. The hotel management agreed with therequest, and identified the supervisors in some FLEs.

The analysis of the kind of activities and skills needed totrain the AI confirmed this need. The API was in fact very userfriendly and did not require someone with a knowledge ofcomputer science, but rather with a wide knowledge of thehotel services, the destination, the needs and tone of languagepreferred by the clients and defined by the brand, etc.Managers, developers and researchers agreed at that point thatthose skills and knowledge were typical of the FLEs, and thatsomeone among them – in particular the more flexible, open-minded and technology-ready FLEs - should be given the taskto supervise the AI training and update it, in order to allow theAI and the robot to correctly interact with the clients. 4 super-visors were identified by the hotel management, one per eachof the languages the AI has to use plus the FO manager. It isimportant to underline that none of the people involved had asupervising role of any kind before being named “AI supervi-sor”, apart from the FO manager, and that the interaction ofthese people with technology until that moment was limited tothe use of the desk computer, of basic suites and of the PMS.

All the AI supervisors received a specific training for thenew tasks, which also encouraged them to observe thecustomers-robot interaction whenever possible, in order toimprove the information level and solve or prevent possiblemisunderstanding in the human-machine interactions.Another specific manual for the “AI supervisors” was de-signed and written, and included instructions on the use of

E. Mingotto et al.

API and on the number of times per week they were to workon it (three-four times a week in the running-in phase and thenat least once a week). A specific tool (Trello TM) and a pro-cedure on how to communicate the updating of informationwere also defined.

As a consequence, the FLEs acting as “AI supervisor”qualify as enablers, although not only towards the customers,but also towards the technology. By training it, and keeping itup to date, the supervisors support the technology in under-standing clients, learning about their needs, and interactingwith them. This role is specifically related to a machine thatlearns how to interact and changes accordingly by adapting tohumans. With all the traditional technologies, instead, oncethe mechanism has been defined by the developers, the ma-chine is not easily modifiable and the user must adapt to it. Inthis case, it is only possible to enable people to understand thetechnology, while in the case of AI the ability required is thatof enabling the technology to understand the people it workswith.

Innovator

The researchers had assumed that the time and attentionresources set free by the introduction of AI could be di-rected towards the role of innovator. However, in the twomonths following the robot introduction, the corporate andhotel management seemed reluctant to explore and evenless recognise this role, so no change was made in the jobdescription or in the tasks assigned to FLEs. One the onehand, the management preferred that FLEs focussed onback-office activities; on the other, given the pioneeringnature of the project, the management was adapting to adevelopment of the FLEs role that was not entirely ex-pected. Indeed, if the role of enabler to support clientswas foreseen, the commitment required by the role ofenabler towards technology was difficult to be anticipated,since the effectiveness of the traditional technologies useduntil that moment (not for service encounters though, asalready said) depended entirely from the IT developers.However, during interviews and meetings one month afterthe robot introduction, the new AI supervisors showed a“natural” inclination in proposing improvements to supportthe clients-AI interaction and more generally the cus-tomers’ relationships with various devices provided bythe resort. As said, only the FO mangers, among the“AI supervisors”, already had a responsibility role.However, the interviews conducted before introducing therobot showed that, beforehand, not even the FO managerhad a role of innovator with respect to technology ingeneral, and, to the inexistent service encounters technol-ogy in particular: of this the chain staff and the hotel GMwere in charge.

Differentiator

What discussed about the innovator role applies to thedifferentiator as well. Although encouraged to reflect on it,the managers were prone to delay this transformation.Differently from what happened with the innovation aspects,however, also the AI supervisors did not advance any propos-al on how they could change their behaviour and skills tomake the difference for the client.

Coordinator

The management again was not ready to recognise this role.The AI supervisors, however, although not formally, assumedthe role of coordinators as a “spin-off” of their role of tech-nology enablers. By deciding which information has to beprovided to guests by the robot and which information byFLEs (e.g. can the clients be informed about the golf courseprices by the robot or do they need to go to the FO, or directlyto the golf course personnel; etc.), the supervisors de factodirect the clients to this or that area for a specific serviceencounter, define when this encounters should be technologysupported or not, and partially instruct the level of co-production of the client. Although the information to be in-cluded in the AI and the way in which to answer some ques-tions from the customer are agreed with the GM and othermanagers, the supervisor has the possibility to suggest anddiscuss coordination aspects not allowed to other FLEs, e.g.in the specific case the supervisor discussed at different levelthe opportunity the AI provides a full information on someservices or it answers to ask directly at the reception desk.Favouring one solution over the other one depends on thecomplexity and update rate of the information and on the factthat the robot could be muchmore effective in providing somehints, having the possibility to display on its screen maps,website, etc. The decision regarding what kind of informationwas to be provided on the printed materials is not only anexample of how the supervisor contributed in trying to im-prove the customer experience, but also of how he contributedin the design of service encounters.

The changing role of customers

Regarding the third research question, findings from the studyshowed that also the transformed role taken by customersqualifies as enabler of the technology. First of all, the encoun-ter provided by the robot can only happen through an activeinput from the guests (as underlined by Larivière et al. (2017),who can decide whether to ask to the robot or to a humanemployee and – if they opt for the robot – who have to enablethe interaction, by selecting the language on the tablet (amongthe guests observed, 82.30% select the language and start to

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

interact with the robot, while the others just watch it, withoutenabling the conversation).

Customers become enabler of the technology also in thesense that they actively contribute to the AI training.However, guests became fully aware of this role, only whenFLEs and the information panels explained to them that theircontribution in interacting with the machine was precious forimproving the AI effectiveness and then the user experience.According to the interviews, customers seem to have positive-ly accepted this role, since all respondents were eager to ac-tively “cooperate”.

Due to the limited period in which the effects of the robotadoption on customers’ experience were observed, findingsare not sufficient to clearly identify other possible changingroles taken by customers. However, we may assume that cus-tomers could act also as a kind of differentiator of the serviceencounter, since, by interacting with the machine and activelygiving inputs for the AI training, they may progressivelyshape and customise the information environment of the AIaccording to their specific requests and needs. They may con-tribute to differentiate the AI from its “initial version” and ingeneral from a more standardized solution that also other sim-ilar competitors could adopt.

Discussion and conclusion

The merit of this study largely lies in the fact that it is based ona concrete real life case and on the action research methodol-ogy, that have allowed the authors to directly observe andassist the company for a whole year, investigating what itmeans to really incorporate an AI-based technology within aservice business environment. The adoption of a robot/AI wasnot conceived as an experiment but as a real asset for thebusinesses involved, showing strategic and operational issues.The project developed thanks to a strong cooperation betweenthe researchers, the software developers, hotel managers andemployers and gave the researchers the opportunity to monitorthe whole process and the critical aspects.

Contribution and theoretical implications

The first contribution of this paper to the studies on servicerobots and AI in the hospitality and service industry is to fillthe gap in empirical research focussed on companies, staff,and organizational re-design, and based on research methods- such as action research - with a strict cooperation with thecompanies themselves (Ivanov et al. 2019). It explores theimpact of robots and AI on operations management of hotelcompanies and specifically how robot and AI change a part ofthese management procedures. Both questions have beenpointed out for future research by Ivanov and Webster(2019a, 2019b, 2019c, 2019d).

Secondly, the case here investigated is a hospitality busi-ness that can be considered “representative” of a quite largepart of businesses of the hospitality sector. Not a “champion”of cutting edge technologies or an experiment of a big inter-national chain, with easy access to significant financial andhuman resources, was chosen. This is a small hotel chain, withno more than 13 hotels and a quite experience in the commoncomputer technologies for hospitality management andmarketing.

Thirdly, the paper focuses on a situation where a high levelof automated social presence combines with a similarly highlevel of human social presence, a configuration still not muchinvestigated (van Doorn et al. 2017). The introduction of AI-based technology within a service business environmentseems not only to change roles, but also to create new rolesboth for customers and employees. In addition, it seems toneed new approaches to be effectively introduced, as theychallenge knowledge and practice accumulated with pasttechnologies.

More specifically, in reference to the first research question- which role these technologies act in the organization: aug-mentation, substitution or network facilitator - the presentstudy shows that the conversational agent adopted in the resortcan qualify as augmentation force, supporting FLEs inassisting guests (Ivanov and Webster 2019d; De Keyseret al. 2019). This role was assigned by the managers whenthey decided to adopt the technology and the AI was trainedaccordingly, but it was not taken for granted since it needed tobe first “accepted” by FLEs and customers. Once the technol-ogy was well trained, the interviews showed how the FLEsstarted to recognise the robot as a support that could relievethem from the task of answering repetitive questions, thusallowing them to dedicate themselves to more complex as-pects of their jobs.

Also the customers, once supported and “educated”,recognised and accepted what the tasks of the robot were; thisconfirms other findings in the literature, which showed thatinformation provision is one of the tasks undertaken by robotsthat has greater customers’ approval (Ivanov and Webster2019a, 2019b; Ivanov et al. 2018).

However, the effort in training the AI and re-designing theroles of employees and customers around it suggests that it isnot yet possible for the robot/AI to substitute completely FLEsin a job requiring the management of complex informationflows and relationships. The first, less “guided”, interactionsof the customers with the robot confirm the tendency to an-thropomorphize it, expecting human abilities - as suggested byvan Doorn et al. (2017), and to actively engage in conversa-tion with the robot to develop a certain level of “relationship”– as suggested in Tung and Au (2018). However, these levelsof interaction require the AI the flexibility to span from one tothe other cognitive domain and relationship kind - a flexibilitythat the most used narrow AI kind is still far from achieving.

E. Mingotto et al.

The second research question addressed the role of front-line employees: a) which changing roles can they take amongenabler, innovator, coordinator and differentiator? b) towardswhom: the customer only, and/or technology? From the theo-retical point of view, this study highlights the existence of arole for FLEs that, although hypothesized within this study,was not foreseen or sufficiently investigated in the existingliterature on the evolving role of FLEs in front of the infusionof technology in service encounters: the role of the AI super-visor (Bolton et al. 2018; De Keyser et al. 2019; Larivièreet al. 2017). This is the most important finding emerging fromthis study, as it has implications on the other FLEs roles and itis a consequence of the presence of a learning technologysupposed to interact with the customer in a natural languageand way, providing a service experience where customers areengaged by technology on a social level (van Doorn et al.2017). The supervisor is essential in ensuring a smooth andeffective service encounter between the robot and the client,but a poor supervision and training of personnel is a high riskboth for the technology itself and for the overall success of itsintroduction.

Not only does the study introduce this role, but, in partic-ular for FLEs, it also suggests how the “AI supervisor” mightbe a key role to access those of coordinator, innovator, etc.Indeed, the role of enabler towards the customer (Larivièreet al. 2017) is fully confirmed for all the FLEs by this case,as the role was formally recognised and the procedureschanged. The change of the procedure and the revision ofthe manual gave a chance also to explore how the serviceoperations manual have been/can be updated as a conse-quence of the “robotisation” of some FO activities (Ivanov2019).

The other roles of FLEs envisaged in the literature - inno-vator, coordinator and differentiator - seem however mainlyaccessible to the “AI supervisor”. There was no element indi-cating that the situation and tasks of the other FLEs mightchange further, since, having “merely” a role of enabler forthe customers, they were simply required to acquire somebasic knowledge about the robot/AI, and to follow a specificprotocol during the check in.

At the same time, a couple of these roles, namely coordi-nator and innovator, emerged spontaneously, as the AI super-visor, being put in full control of the information and of the AI,was able to suggest how to use it for a better tourist experi-ence, work and information flow, although the managementdecided not to recognise these other roles. However, the studypoints out how the fostering of those new roles is not so easyon the management’s part. Indeed, Bowen (2016) commentshow the coordinator role is the most important, but the lessrecognized by the managers. The authors are presently study-ing other cases of conversational agents introduced to supportand foster FO and booking/ordering activity in the hospitalityindustry and restaurants, which seem to confirm the key im-portance of the “AI supervisor”.

The third research question was what changing roles cus-tomers can take among enabler, innovator, coordinator anddifferentiator. The study provides evidence about the chang-ing role of customers, who act as enabler of the technology,not only because it is up to them to give the input to themachine and start the interaction, but also because they active-ly contribute to the AI machine learning. However, customersbecame full enablers only after they received information andguidance by the FLEs about the robot functions (it as an as-sistant and not a gadget or a toy to entertain guests) and the

Fig. 2 The role of technology andthe main changing roles of FLEsand customers

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

way the AI learns. Indeed, once customers had been “educat-ed”, they started to interact with the machine as expected byasking questions that were consistent with the tasks the AI wastrained for and then by supporting its machine learning. Thecustomer-employee rapport building is important for mediat-ing the relationship between the robot and the customers, assuggested by Qiu et al. (2019).

This accounted role for the customer, as the AI supervisorone, was not foreseen in existing literature (Bolton et al. 2018;De Keyser et al. 2019; Larivière et al. 2017), which does notappear to consider that learning machines are evolving ones,and thus need interaction to do so, differently from the tech-nologies so far known, which, once debugged, are expected towork perfectly from the first interaction.

See Fig. 2 and Table 2 for a synthesis.

Managerial implications

From the managerial point of view, the study highlights thatthe success of this kind of automation does not depend only onthe selection of the solution (robots and AI) that best fits thecompany’s profile and operating environment. It also stronglydepends on the assimilation process (Hasgall and Ahituv2018) and then on efforts made by the company to re-organise some operations and to re-design customers’ andFLEs’ roles, also by providing a differentiated training andguidance on how to use, interact and manage the AI system.

The study indicates, however, that managers as well asacademics operate on the basis of previous experiences withtechnologies that, once installed, are fully developed, wherethe employees or customers are supposed to act within alreadydefined steps. This may lead them to underestimate howmuchthe machine learning systems need in terms of training andattention, and how the employment of these systems requires are-definition of the employees’ role, not only in order to findspaces the technology does not occupy but also simply toallow the system to properly work. This implies that the man-agers need to be aware of the full implications of the AI adop-tion in terms of service design, ecosystem, employees’ skillsand new coordinating needs.

For the managers it might also not be as easy as it seems torethink FLEs in roles that are far from traditional ones andrequire specific training. Furthermore, to full exploit the po-tential of innovation, differentiation, etc. offered by the AI, themanagers should probably change the job description and therecruitment criteria. A similar situation might be found inthose hotels where the FO desk was cancelled and FLEs weretransformed into guest experience executives, who were re-quired to stay in the lobby and interact with the customer: thecustomers were lost in the lobby as they did not know whatthey were supposed to do and who they should look for, whilethe employees were too used to wait for the customers, and sotended not to approach them, or had difficulties to tell new

customers from others already checked in (Accor Group andSkift 2019). In this case too customers and employees neededto be instructed and trained for their new roles.

As suggested by Larivière et al. (2017), the study givesevidence that both employees and customers need to updatetheir skills or gain new ones to assume the new roles. From thecustomer-side perspective, the development of new skillscould seem not so challenging, since social robots and con-versational systems based on AI are conceived and designedin order to interact in a user-friendly and socially-compatibleway and to understand questions posed in natural language.However, for ensuring a successful interaction, it is essentialthat clear information is provided to customers not only aboutthe tasks of the machine (what use the machine is meant forand what it is able to answer) but also about how importantcustomers themselves are in contributing to the AI trainingand development. As suggested by Ciuchita et al. (2019),customers have to be supported in coping with the “innova-tion”, in order to make them more eager and maximise theirgains in using the machine.

For FLEs, and in particular for the AI supervisors, specificknowledge and skills have to be provided about the technol-ogy adopted in the business (kind of AI and how it works,mechanisms of the supervised machine-learning, robot fea-tures, management of the information provided by the ma-chine, etc.) and the main issues related to the interaction withcustomers. An in-depth knowledge about the organisation andits processes is a pre-requisite condition; this is why it is betterto find the AI supervisors among the FO staff, paying howeverparticular attention to those employees with both technologyreadiness and soft skills (as supposed by Larivière et al. 2017),such as open-mindedness and flexibility.

Limitations

The present research is an exploratory study based on a spe-cific case and on a project in which the authors were engagedas consultants for the businesses. As a consequence, the co-operation with the researchers might be considered a limita-tion, as the software developer and the hotel managers andstaff, left to their own devices, would have possiblyapproached the training of the AI and its introduction in adifferent way. However, needs in terms of procedures, roles,ad skills emerged independently from the researchers, and thefact that some roles were rejected when addressed (innovator,coordinator, etc.) shows how the other partners were not soinfluenced in their decision making by the researchers’presence.

Another limitation is linked to the fact that the authors,partially also because of a hardware problem that requiredthe robot to be sent to the producer to be repaired, were notable to study the AI introduction for a longer period.

E. Mingotto et al.

Table2

FLEs’roles–Beforeandaftertherobotintroduction

Rolewith

respectto

service

encounters/customer

interactionspecific

technology

BeforeAI/robotintroduction

After

AI/robotintroduction

FO

employee

-supervisor

Ena

bler

Not

partof

thejobdescriptionandactual

procedures

Enabler

towards

thecustom

ers:1)

changesin

procedures

regardingwhattosayat

thewelcomingin

orderto

highlig

htthebasics

ofinteractingwith

therobot/A

I;2)

developm

enta

specifichandbook

ontheAIfeatures

andabilitiesintroduced

asapartof

theFO

operations.

Enabler

towards

technology

–training

theAIto

betterunderstand/in

teractwith

the

custom

er:1

)Specificroleof

AIsupervisor

introduced

andintegrated

insome

employeesactiv

ities;2

)developm

ento

faspecifichandbook

onhowto

traintheAI;

3)aspecifictraining

fortherole.

Inno

vator

Not

partof

thejobdescriptionandactual

procedures

Not

introduced

although

suggestedduring

theactio

nresearch.A

Isupervisors

suggestedsomeim

provem

entintheinform

ationchannels

Nospecifictraining

Differentiator

Not

partof

thejobdescriptionandactual

procedures

Not

introduced

although

suggestedduring

theactio

nresearch

Coordinator

Not

partof

thejobdescriptionandactual

procedures

Partially

developed:

naturally

took

bythesupervisorson

asaconsequence

ofbeingenablerforboth

technology

andcustom

ers,butn

otfully

recognised.

Nospecifictraining

FO

employees–no

t“sup

ervisor”

Ena

bler

Not

partof

thejobdescriptionandactual

procedures–

Enabler

towards

thecustom

ers:1)

changesin

procedures

regardingwhatto

sayatthewelcomingin

orderto

highlig

htthebasics

ofinteractingwith

therobot/A

I;2)

developm

enta

specifichandbook

ontheAIfeatures

andabilitiesintroduced

asapartof

the

FOoperations.

Inno

vator

Not

partof

thejobdescriptionandactual

procedures

Not

immediately

introduced

orencouraged

Differentiator

Coordinator

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

Another issue is that, given the nature of the project, therewas the need to balance the main practical needs and requestsof the businesses with the research purposes and design. Thisaspect may be considered a limitation, for example in relationto the low number of interviews to customers, since businessmanagers were not so inclined to disturb guests during theirstay. In addition, in order to assess the AI evolving impact oncustomers’ perception and satisfaction, respondents should bethe same at the beginning and at the end of the monitoringperiod. Unfortunately, it was impossible to interview the samecustomers since respondents were guests who were staying inthe hotel at the time the interviews were done and who nevercame back again during the two-month monitoring period.

In addition, the results of this study need to be validatedthrough further research on other cases, possibly consideringdifferences between large international hotel chains and SMEsand other kind of conversational agents. The very reason thatled the decision to adopt this technology in this specific caseof a large resort may not have sense in smaller hotels; also, theway of approaching the changes brought by this machine maydiffer if other typologies of conversational agents areimplemented.

Future research

Although the project allowed an almost unique access to dataand observation, it is essential to analyse what happens incases with no direct support form researchers. The growingof some FLEs into the role of “AI supervisor” that this papercontributes to introduce and integrate to the existing studiesneeds to be confirmed to exist in other scenarios and environ-ments, also in order to understand if it is always a responsibil-ity of FLEs, or if different choices are possible and what theseentail for the employees themselves and the management.

On this note, a first analysis conducted by the authors onother cases seems to indicate that, to avoid the technologyfailure, the software developer might decide to continue thetraining of the AI for as long as it works, recruiting staffspecifically for this purpose, while FLEs, not involved direct-ly, keep on with their usual tasks. This kind of scenario wouldthen lead the “AI supervisor” role to be externalised. Wouldthat encourage the managers and the software developers tothink of a technology that is a substitution instead of an aug-mentat ion force? And would that mean a strongstandardisation of the service encounters?

Future research should expand considering also the impactdetermined by other AI devices and not only physical human-oid robots. Indeed, with respect to the customer role, it wouldbe essential to understand if customers’ approach and interac-tion would be different in front of a different support (e.g.chatbot), and if, in this case, their role as enabler, innovator,etc. would be more or less fully developed. Another factor tobe explored is the role of customers and employees if the AI

supports service encounters that happen before the arrival ofthe customer at the hotel/destination. In addition, further re-search should investigate the cases in which multiple devicesand conversational agents are adopted in the service encounterat different levels of the customer journey (before booking,onsite, etc.), exploring the relations among devices, and be-tween customers, agents and employees.

Funding Open Access funding provided by Università Ca' FoscariVenezia within the CRUI-CARE Agreement.

Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing, adap-tation, distribution and reproduction in any medium or format, as long asyou give appropriate credit to the original author(s) and the source, pro-vide a link to the Creative Commons licence, and indicate if changes weremade. The images or other third party material in this article are includedin the article's Creative Commons licence, unless indicated otherwise in acredit line to the material. If material is not included in the article'sCreative Commons licence and your intended use is not permitted bystatutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of thislicence, visit http://creativecommons.org/licenses/by/4.0/.

References

Accor Group and Skift. (2019). The Transformative Role of Design inHospitality. Retrieved from https://skift.com/2019/06/06/new-skift-report-the-transformative-role-of-design-in-hospitality/.

Aghaei, S., Nematbakhsh, M. A., & Farsani, H. K. (2012). Evolution ofthe world wide web: From WEB 1.0 TO WEB 4.0. InternationalJournal of Web & Semantic Technology, 3(1), 1–9.

Bolton, R. N.,McColl-Kennedy, R. J., Cheung, L., Gallan, A., Orsingher,C., Witell, L., & Zaki, M. (2018). Customer experience challenges:Bringing together digital, physical and social realms. Journal ofService Management, 29(5), 776–808. https://doi.org/10.1108/JOSM-04-2018-0113.

Bowen, D. E. (2016). The changing role of employees in service theoryand practice: An interdisciplinary view. Human ResourceManagement Review, 26(1), 4–13. https://doi.org/10.1016/j.hrmr.2015.09.002.

Buhalis, D., & Main, H. (1998). Information technology in peripheralsmall and medium hospitality enterprises: Strategic analysis andcritical factors. International Journal of Contemporary HospitalityManagement, 10(5), 198–202. https://doi.org/10.1108/09596119810227811.

Ciuchita, R., Mahr, D., & Odekerken-Schröder. (2019). Deal with it:How coping with e-service innovation affects the customer experi-ence. Journal of Business Research, 103, 130–141. https://doi.org/10.1016/j.jbusres.2019.05.036.

Chen, Y., & Hu, H. (2013). Internet of intelligent things and robot as aservice. Simulation Modelling Practice and Theory, 34, 159–171.https://doi.org/10.1016/j.simpat.2012.03.006.

Cobos, M., Mejia, C., Bulent Ozturk, A., & Wang, Y. (2016). A technol-ogy adoption and implementation process in an independent hotelchain. International Journal of Hospitality Management, 57, 93–105. https://doi.org/10.1016/j.ijhm.2016.06.005.

Collins, G. R., Cobanoglu, C., Bilgihan, A., & Berezina, K. (2017).Hospitality information technology: Learning how to use it. (8th

E. Mingotto et al.

ed.). Dubuque, IA: Kendall/hunt publishing co. chapter 12:Automation and robotics in the hospitality industry, 413-449.

Coughlan, P., & Coghlan, D. (2002). Action research for operations man-agement. International Journal of Operations and ProductionManagement, 22(2), 220–240. https://doi.org/10.1108/01443570210417515.

Cruz-Jesus, F., Pinheiro, A., & Oliveira, T. (2019). Understanding CRMadoption stages: Empirical analysis building on the TOE framework.Computers in Industry, 109, 1–13. https://doi.org/10.1016/j.compind.2019.03.007.

De Keyser, A., Köcher, S., Alkire, L., Verbeeck, C., & Kandampully, J.(2019). Frontline service technology infusion: Conceptual arche-types and future research directions. Journal of ServiceManagement, 30(1), 156–183. https://doi.org/10.1108/JOSM-03-2018-0082.

Den Hertog, P., van der Aa, W., & de Jong, M. W. (2010). Capabilitiesfor managing service innovation: Towards a conceptual framework.Journal of Service Management, 21(4), 490–514. https://doi.org/10.1108/09564231011066123.

Dogan S., & Vatan A. (2019). Hotel Managers’ Thoughts Towards NewTechnologies and Service Robots’ at Hotels: A Qualitative Study inTurkey. In Cobanoglu C., Cavusoglu M., & Corbaci A. (Eds.).Advances in Global Business and Economics, 2. 382-399. https://doi.org/10.5038/9781732127555.

Escuer, P., Mateo, A., McConnell, C., & Schutes, J. (2014). Refiningvisitor tracking for museum Victoria interactive. Faculty ofWorcester Polytechnic Institute: Qualifying Project Reportsubmitted.

Hasgall, A., &Ahituv, N. (2018). Implementing continuous adapataion totechnology innovation in complex adaptive organizations. Journalof High Technology Management Research, 29, 35–45. https://doi.org/10.1016/j.hitech.2018.04.004.

Ivanov, S. (2019). Ultimate transformation: How will automation tech-nologies disrupt the travel, tourism and hospitality industries?Zeitschrift für Tourismuswissenschaft, 11(1), 25–43.

Ivanov, S., & Webster, C. (2017). Adoption of robots, artificial intelli-gence and service automation by travel, tourism and hospitalitycompanies - a cost-benefit analysis. In International ScientificConference “Contemporary tourism – traditions and innovations”,19–21 October 2017. Sofia: University.

Ivanov S., Webster C. (2019a) What Should Robots Do? A ComparativeAnalysis of Industry Professionals, Educators and Tourists. In:Pesonen J., Neidhardt J. (eds) Information and CommunicationTechnologies in Tourism 2019. Springer, Cham. https://doi.org/10.1007/978-3-030-05940-8_20.

Ivanov S., Webster C. (2019b) Perceived Appropriateness and Intentionto Use Service Robots in Tourism. In: Pesonen J., Neidhardt J. (eds)Information and Communication Technologies in Tourism 2019.Springer, Cham. https://doi.org/10.1007/978-3-030-05940-8_19.

Ivanov, S., & Webster, C. (2019c). Conceptual framework of the use ofrobots, artificial intelligence and service automation in travel, tour-ism, and hospitality companies. In Ivanov, S., & Webster, C. (Eds.)Robots, Artificial Intelligence and Service Automation in Travel,Tourism and Hospitality. 7-37. https://doi.org/10.1108/978-1-78756-687-320191001.

Ivanov, S., & Webster, C. (2019d). Economic fundamentals of the use ofrobots, artificial intelligence and service automation in travel, tour-ism and hospitality. In Ivanov, S., & Webster, C. (Eds.) (2019)Robots, Artificial Intelligence and Service Automation in Travel,Tourism and Hospitality. 39-55. https://doi.org/10.1108/978-1-78756-687-320191002.

Ivanov, S., Webster, C., & Berezina, K. (2017). Adoption of robots andservice automation by tourism and hospitality companies. RevistaTurismo & Desenvolvimento, 27(28), 1501–1517.

Ivanov, S., Webster, C., & Seyyedi, P. (2018). Consumers' attitudes to-wards the introduction of robots in accommodation establishments.Tourism, 63(3), 302–317.

Ivanov, S., Gretzel, U., Berezina, K., Sigala, M., & Webster, C. (2019).Progress on robotics in hospitality and tourism: A review of theliterature. Journal of Hospitality and Tourism Technology, 10(4),489–521. https://doi.org/10.1108/JHTT-08-2018-0087.

Järvinen, J., & Taiminen, H. (2016). Harnessing marketing automationfor B2B content marketing. Industrial Marketing Management, 54,164–175. https://doi.org/10.1016/j.indmarman.2015.07.002.

Kazandzhieva, V., & Filipova, H. (2019). Customer attitudes towardsrobots in travel, tourism, and hospitality: A conceptual framework.In S., Ivanov, & C., Webster (eds). Robots, artificial intelligence andservice automation in travel, tourism and hospitality. EmeraldPublishing. https://doi.org/10.1108/978-1-78756-687-320191004.

Kuo, C., Chen, L., & Tseng, C. (2017). Investigating an innovative ser-vice with hospitality robots. International Journal of ContemporaryHospitality Management, 29(5), 1305–1321. https://doi.org/10.1108/IJCHM-08-2015-0414.

Lam, C., & Law, R. (2019). Readiness of up-scale and luxury brandedhotels for digital transformation. International Journal ofHospitality Management, 79, 60–69. https://doi.org/10.1016/j.ijhm.2018.12.015.

Larivière, B., Bowen, D., Andreassen, T. W., Kunz, W., Sirianni, N. J.,Voss, C., Wünderlich, N. V., & De Keyser, A. (2017). Serviceencounter 2.0′: An investigation into the roles of technology, em-ployees and customers. Journal of Business Research, 79, 238–246.https://doi.org/10.1016/j.jbusres.2017.03.008.

Li, J., Bonn, M., & Ye, B. (2019). Hotel employee's artificial intelligenceand robotics awareness and its impact on turnover intention: Themoderating roles of perceived organizational support and competi-tive psychological climate. Tourism Management, 73, 172–181.https://doi.org/10.1016/j.tourman.2019.02.006.

MacDonald, C. (2012). Understanding participatory action research: Aqualitative research methodology option. The Canadian Journal ofAction Research, 13(2), 34–50. https://doi.org/10.33524/cjar.v13i2.37.

Makridakis, S. (2017). The forthcoming artificial intelligence (AI) revo-lution: Its impact on society and firms. Futures, 3, 1–27. https://doi.org/10.1016/j.futures.2017.03.006.

Marinova, D., de Ruyter, K., Huang, M.-H., Meuter, M. L., &Challagalla, G. (2017). Getting smart: Learning from technology-empowered frontline interactions. Journal of Service Research,20(1), 29–42. https://doi.org/10.1177/2F1094670516679273.

McCarthy, J. (2007). What is artificial intelligence? http://jmc.stanford.edu/articles/whatisai/whatisai.pdf.

Mori, M. (1970). The uncanny valley. Energy, 4(7), 33–35.Murphy, D. (2018). Silver bullet ormillstone? A review of success factors

for implementation of marketing automation. Cogent Business &Management, 5, 1–10. https://doi.org/10.1080/23311975.2018.1546416.

Murphy, J., Hofacker, C., & Gretzel, U. (2017). Dawning of the age ofrobots in hospitality and tourism: Challenges for teaching and re-search. European Journal of Tourism Research, 15, 104–111.

Murphy, J., Gretzel, U., & Pesonen, J. (2019).Marketing robot services inhospitality and tourism: The role of anthropomorphism. Journal ofTravel & Tourism Marketing, 36(7), 784–795. https://doi.org/10.1080/10548408.2019.1571983.

Panch, T., Szolovitz, P., & Atun, R. (2018). Artificial intelligence, ma-chine learning and health systems. Journal of Global Health, 8(2).https://doi.org/10.7189/2Fjogh.08.020303.

Pinillos, R., Marcosa, S., Feliz, R., Zalamab, E., & Gómez-García-Bermejo, J. (2016). Long-term assessment of a service robot in ahotel environment. Robotics and Autonomous Systems, 79, 40–57.https://doi.org/10.1016/j.robot.2016.01.014.

Challenges in re-designing operations and jobs to embody AI and robotics in services. Findings from a case...

Qiu, H., Li, M., Shu, B., & Bai, B. (2019). Enhancing hospitality expe-rience with service robots: The mediating role of rapport building.Journal of Hospitality Marketing & Management, 29(3), 1–22.https://doi.org/10.1080/19368623.2019.1645073.

Speier, C., & Venkatesh, V. (2002). The hidden minefields in the adop-tion of sales force automation technologies. Journal of Marketing,66, 98–111. https://doi.org/10.1509/2Fjmkg.66.3.98.18510.

Tung, V. W. S., & Au, N. (2018). Exploring customer experiences withrobotics in hospitality. International Journal of ContemporaryHospitality Management, 30(7), 2680–2697. https://doi.org/10.1108/IJCHM-06-2017-0322.

Tung, V. W. S., & Law, R. (2017). The potential for tourism and hospi-tality experience research in human-robot interactions. InternationalJournal of Contemporary Hospitality Management, 29(10), 2498–2513. https://doi.org/10.1108/IJCHM-09-2016-0520.

Tussyadiah, I. (2020). A review of research into automation in tourism:Launching the annals of tourism research curated collection on arti-ficial intelligence and robotics in tourism. Annals of TourismResearch, 81(102883), 1–13. https://doi.org/10.1016/j.annals.2020.102883.

Tussyadiah, I. P., & Park, S. (2018). Consumer evaluation of hotel servicerobots. In B. Stangl & J. Pesonen (Eds.), Information and

communication Technologies in Tourism (pp. 308–320). Cham:Springer. https://doi.org/10.1007/978-3-319-72923-7_24.

van Doorn, J., Mende, M., Noble, S. M., Hulland, J., Ostrom, A. L.,Grewal, D., & Petersen, J. A. (2017). Domo arigato Mr. Roboto:Emergence of automated social presence in organizational frontlinesand Customers' Service experiences. Journal of Service Research,20(1), 43–58. https://doi.org/10.1177/2F1094670516679272.

Wirtz, J., Patterson, P. G., Kunz, W. H., Gruber, T., Nhat Lu, V., Paluch,S., & Martins, A. (2018). Brave new world: Service robots in thefrontline. Journal of Service Management, 29(5), 907–931. https://doi.org/10.1108/JOSM-04-2018-0119.

Yalowitz, S. S., & Bronnenkant, K. (2009). Timing and tracking:Unlocking visitor behavior. Visitor Studies, 12(1), 47–64. https://doi.org/10.1080/10645570902769134.

Yu, C. E. (2019). Humanlike robots as employees in the hotel industry:Thematic content analysis of online reviews. Journal of HospitalityMarketing & Management, 29(1), 1–17. https://doi.org/10.1080/19368623.2019.1592733.

Publisher’s note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

E. Mingotto et al.